an aerobic exercise training program on brain function of sedentary older people.
... jects could correctly report the orientation, left or right, of the line [27].
Neurobiology of Aging, Vol. 5, pp. 35-42, 1984. ©AnkhoInternationalInc. Printed in the U.S.A.
0197-4580/84$3.00 + .00
Aerobic Exercise Training and Improved Neuropsychological Function of Older Individuals I R O B E R T E. D U S T M A N , 2"a'4 R O B E R T O. R U H L I N G , 5 E W A N M. R U S S E L L , 6 D O N A L D E. S H E A R E R , 2 H. W I L L I A M B O N E K A T , ~'7 J O H N W . S H I G E O K A , 2'7 J A M E S S. W O O D ~'r A N D D A V I D C. B R A D F O R D 4
Neuropsychology Research Laboratory (151A), VA Medical Center 500 Foothill Blvd., Salt Lake City, UT 84148 R e c e i v e d 19 S e p t e m b e r 1983 DUSTMAN, R. E., R. O. RUHLING, E. M. RUSSELL, D. E. SHEARER, H. W. BONEKAT, J. W. SHIGEOKA, J. S. WOOD AND D. C. BRADFORD. Aerobic exercise training and improved neuropsychologicalfanction of older individuals. NEUROBIOL AGING 5(1) 35-42, 1984.--The effects of a four month aerobic exercise conditioning program on neuropsychological test performance, depression indices, sensory thresholds, and visual acuity of 55-70 year old sedentary individuals were evaluated. Aerobically trained subjects were compared with two age-matched control groups of subjects: those who trained with strength and flexibility exercises and others who were not engaged in a supervised exercise program. The aerobically trained subjects demonstrated significantly greater improvement on the neuropsychological test battery than did either control group. Depression scores, sensory thresholds, and visual acuity were not changed by aerobic exercise. The pattern of results suggests that the effect of aerobic exercise training was on central rather than on peripheral function. We speculate that aerobic exercise promoted increased cerebral metabolic activity with a resultant improvement in neuropsychological test scores. Aerobic exercise Response time
Aging CFF Sensory thresholds
Depression
Digit span
GROWING old is accompanied by a gradual decline of the central nervous system (CNS). Measures of higher mental function such as intellect, memory, attention, and perception evidence decline [12] and behavior slows as demonstrated by prolonged reaction times [12], reduced brain wave (EEG) frequency [51], increased latency of event related potentials [6,25], and slower nerve conduction velocities [24]. It has been suggested that decrements in mental and electrophysiological functioning of older individuals may, in part, result from the brain being mildly hypoxic [33,47]. There are two factors which contribute to reduced cerebral oxygenation in old age and thus may adversely affect brain function: the increasing presence of atherosclerosis [8,49] and an inability to efficiently transport and utilize oxygen resulting from physically inactive life-styles [22]. The latter can be improved by aerobic exercise [22] and there is growing evidence suggesting that the rate of decline of physical and cognitive abilities is governed by physical conditioning
Digit symbol
Human
Memory
level as well as by age [ 10, 11,23, 32]. For example, response times of older men who had maintained an active participation in physical activities such as racquet sports and running were significantly faster than those of age-matched sedentary men and little different from response times of much younger sedentary subjects [60,64]. Also, highly fit older individuals scored higher on tests of fluid intelligence than did less fit subjects [28,56]. Results such as these raise important questions. Does the better performance of physically active older individuals reflect a predisposition for superiority in both athletic and cognitive abilities, or does exercise per se have a beneficial effect on CNS functioning? If the latter is true, can CNS functioning of older people be significantly improved by a program of physical activity even though they have maintained a sedentary life-style for many years? Is the type of exercise important? Exercise that results in increased aerobic efficiency, i.e., an improved ability to transport oxy-
1This research was supported by the Veterans Administration and by funds from NIH Biomedical Research Support (Grant No. RR07092). zVeterans Administration Medical Center, Salt Lake City. 3Department of Neurology, University of Utah. 4Department of Psychology, University of Utah. ~College of Health, University of Utah. 6College of Health, Wheaton College. 7Department of Internal Medicine, University of Utah.
35
DUSTMAN ET AL.
36 gen from the environment to consumer cells [23], may have more effect on brain function than physical activities that do not improve aerobic capacity. The present study was designed to evaluate the effects of an aerobic exercise training program on brain function of sedentary older people. METHOD Sedentary individuals aged 55-70 years were solicited from the community and screened for health problems which would preclude their participation in an exercise program. Those who stated that they actively engaged in physical conditioning activities were not considered further. The research was described in greater detail at a formal meeting and prospective subjects were provided an opportunity to ask questions and become familiar with a treadmill on which maximal exercise tests were to be performed. They were told they would be paid a modest sum at the end of the study for their participation and were given informed consent forms to review. The protocol for this study and the consent forms were approved by the University of Utah Review of Research with Human Subjects Committee (IRB). Individuals who elected to participate in the research were scheduled for a maximal exercise test. Immediately before this test they were examined by a physician for health problems which would exclude them from safely performing a maximal exercise test and/or engaging in an exercise program. Those selected to participate performed a modified Balke exercise test on a motor driven treadmill [2]. During the test their electrocardiogram was continuously monitored by a physician and their blood pressure was measured every two to three minutes. The treadmill was set at a speed of 67.0 m/min and 1% slope. While the speed remained constant, the slope of the treadmill was increased 1% each succeeding minute of exercise until a maximal effort was achieved [1]. During the test a standard open-circuit indirect calorimetry system was used so that measures of minute ventilation and maximal oxygen uptake (VOzmax) could be obtained. On completion of the maximal exercise test, subjects were alternately assigned to the experimental group (aerobic exercise training) or to an exercise control group. A questionnaire revealed that only three subjects smoked on a regular basis; these were in the aerobic exercise group. Additional measures were obtained during two test sessions, each about 90 minutes in duration, from subjects in the experimental and exercise control groups and from a third group of older volunteers. The latter, nonexercise controls, did not participate in an exercise program and were not tested on the treadmill. They were screened for health problems during a structured interview. Electrophysiological measures (EEG and evoked potentials), auditory, visual and somatosensory thresholds, and a measure of visual acuity were obtained during one test session. During the other, measures were obtained for several neuropsychological tests and two depression inventories, the Beck Depression Index [4] and the Self-Rating Depression Scale [69]. The EEG and evoked potential results will be reported elsewhere. All subjects reported they were right-handed.
Sensory Thresholds and Visual Acuity As part of the procedures for recording auditory brain stem potentials, an auditory threshold was measured. Both ears were tested; the best ear was reported. Clicks were generated by a Grass Auditory Stimulus Control Module
(S10ASCM) and delivered monaurally to subjects via earphones. Somatosensory thresholds were established for 0.5 msec shocks generated by a Grass S10SCM stimulator and delivered to the median nerve of the dominant hand. Flashes, generated by a Grass PS22 Photostimulator, backlighted a narrow black diagonal line placed in a viewing box towards which the subject's gaze was oriented. Neutral density filters were used to change stimulus intensity. Visual threshold was defined as the lowest intensity at which subjects could correctly report the orientation, left or right, of the line [27]. Visual acuity was measured with a Bausch and Lomb Vision Tester (Model 712241). Vision of a number of subjects was corrected by glasses which were worn during testing procedures.
Neuropsychological Tests (l) Critical Flicker Fusion Threshold (CFF) was measured with a Lafayette Instrument Flicker Fusion Control Unit (Model 12025) with attached viewing chamber (Model 12026) which presented a flashing light to the dominant eye. The light/dark ratio of the stimuli was l : l . CFF threshold was the frequency (Hz) at which the flashes appeared to fuse into a continuous light. (2) Culture-Fair Intelligence, scale 3, Form A, consisting of four timed paper and pencil tests was used as a measure of intellectual performance [48]. (3) Digit Span, a subtest from the Wechsler Adult Intelligence Scale (WAIS) [68], provided a measure of recent memory. The subject's score was the number of digits in the longest series of numbers he/she could correctly repeat plus the number in the longest series which he/she could correctly report in reverse order. (4) Digit Symbol WAIS subtest [68]. Subjects were asked to match numbers with appropriate symbols to be drawn below the numbers. A key illustrating numbersymbol matches was provided. Their score was the number of matches correctly made in 90 sec. (5) Dots Estimation. Sixteen slides, each containing from 1 to 16 opaque dots, were presented tachistoscopically on a screen for 200 msec. Subjects were asked to estimate the number of dots displayed with the score being the number of errors (difference between his/her estimate and the actual number of dots) averaged over two repetitions. (6) Reaction Time. Measures of simple and choice reaction time were obtained. Subjects sat facing a video screen while holding a response switch in each hand and were instructed to respond to appropriate stimuli as quickly as possible. For simple reaction time the imperative stimulus was an X. During choice reaction time trials an X and an O were presented simultaneously, one on the left and the other on the right of the screen. Subjects responded with their left switch when the X appeared to their left and with the fight switch when the X was displayed to their right. Imperative stimuli were preceded by a warning stimulus, a small rectangle. The interval between warning and imperative stimuli was randomly varied among 0.50, 0.75, 1.00, and 1.25 sec. Inter-trial intervals were 1 sec. Fifty valid simple and choice reaction time trials were obtained from each subject. Invalid trials were those for which reaction times were less than t00 msec (anticipation), longer than 500 msec, or a wrong switch was depressed. The fastest and slowest five trials were discarded and mean reaction time was computed from the remaining 40 trials. (7) Stroop Color Test. Three 35.5× 10 cm cards were used as stimulus materials. Card 1 consisted of 17 color names printed in black ink. Card 2 consisted of 17 colored bars. On card 3 were 17 color names printed in a different color of ink (e.g., the word " r e d " was
EXERCISE A N D N E U R O P S Y C H O L O G I C A L F U N C T I O N printed in blue ink). Words and color bars, each about 2-3 cm in length, were ordered vertically on the cards. Subjects completed four tasks, in order, as follows. They were asked to read the color words on card 1 (Task I), name the colors on card 2 (Task II), read the color words on card 3 (Task III), and to name the color of the ink used for each color word on card 3 (Task IV). The latter, an "interference" task, provides a measure of a subject's ability to "shift his perceptual set to conform to changing d e m a n d s " ([45], p. 523), and is particularly sensitive to the effects of adult aging [16]. Each task was timed and subjects were asked to work as rapidly as possible. Exercise Protocol The exercise groups met for three one hour sessions a week over a four month period; each was supervised by a graduate student trained in exercise physiology. Every two weeks the instructors alternated groups. Subjects in the experimental group, following a few minutes of " w a r m u p " exercises, then concentrated on aerobic exercise, consisting mostly of fast walking with occasional slow jogging. Their goal was to increase their heart rate to 70--81)% of their heart rate reserve and to maintain it at this rate for longer periods of time as their conditioning improved [1]. The exercise control group participated in strength and flexibility exercises. They also monitored their heart rates but were encouraged to keep them below a level reported to improve aerobic efficiency. On conclusion of the four month exercise program, subjects in the exercise groups again performed the maximal exercise test; the remaining tests were administered to all subjects. The aerobic exercise, exercise control and nonexercise control groups included, respectively, 13 subjects aged 55-68 years (mean=60.6), 15 subjects aged 55-70 years (mean=62.3) and 15 subjects aged 51-70 years (mean=57.4). Nine of the subjects in each group were males. The groups were equivalent in terms of number of years of education ( 15- i 6 years), scores on the Culture Fair Test of Intelligence, and mean number of days between pre- and posttreatment tests (about 140 days). The groups were not equal, however, with respect to age as subjects in the nonexercise control group were younger than those in the exercise control group (p0.10). To determine if our subjects had made a maximal effort on the treadmill exercise test, we compared mean Oz uptake for the minute preceding the final minute of effort (22.1 ml/kg/min) with that for the final minute (22.9 ml/kg/min), a 0.8 ml/kg/min difference. Means were based on pre- and posttest evaluations of all of the exercise subjects. Maximum oxygen uptake is reportedly achieved if the difference be-
37 TABLE 1 PHYSIOLOGICALMEASURES OBTAINED FROM THE AEROBIC AND EXERCISE CONTROL SUBJECTS WHILE AT REST IMMEDIATELY BEFORE THEIR INITIAL TREADMILLTEST (BLOOD PRESSURE AND HEART RATE) AND WHILE THEY WALKED ON THE TREADMILL (MAXIMUMHEART RATE AND MAXIMUM MINUTE VENTILATION [VE] Aerobic
Exercise Control
Mean S.D. Mean S.D. Blood Pressure (mm Hg) Systolic Diastolic Heart Rate (BPM) At Rest Maximum Maximum VE (L/min)
VO2max(ml/kg/min)
140.1 90.7
15.3 135.8 6.1 85.9
9.2 9.1
82.4 1 6 . 3 79.9 1 2 . 5 148.0 12.8 147.5 1 2 . 6 64.9 2 2 . 1 65.2 1 8 . 8 19.4 5.7 22.5 5.1
t-Value
0.882 1.651 0.450 0.110 0.025 1.507
The probability for each t-value was >0.10 (26 dJ).
tween consecutive VOz readings, with increasing work loads, is less than 2.1 ml/kg/min [65]. Our data suggest that this criterion was met. A Group × Session A N O V A was computed on pre- and posttreatment VO2max values for the two exercise groups. A significant treatment effect was found with VO2max means being larger at the conclusion o f the exercise programs, F(1,26)=23.6, p0.10), a significant interaction, F(1,26)=4.89, p =0.036, indicated a differential effect o f type of training on VO2max improvement. VOzmax increase for the experimental group was significantly greater than that for the exercise control subjects as determined by a t-test of VOzmax change, t(26)=2.153, p=0.04. VO2m~x for the aerobically trained group increased by 27%, from 19.4 to 24.6 ml/kg/min, t(12)=4.08, p
